Abstract:This paper studies formal synthesis of controllers for continuous-space systems with unknown dynamics to satisfy requirements expressed as linear temporal logic formulas. Formal abstraction-based synthesis schemes rely on a precise mathematical model of the system to build a finite abstract model, which is then used to design a controller. The abstraction-based schemes are not applicable when the dynamics of the system are unknown. We propose a data-driven approach that computes a growth bound of the system using a finite number of trajectories. The computed growth bound together with the sampled trajectories are then used to construct the abstraction and synthesise a controller. Our approach casts the computation of a growth bound as a robust convex optimisation program (RCP). Since the unknown dynamics appear in the optimisation, we formulate a scenario convex program (SCP) corresponding to the RCP using a finite number of sampled trajectories. We establish a sample complexity result that gives a lower bound for the number of sampled trajectories to guarantee the correctness of the growth bound computed from the SCP with a given confidence. Our sample complexity result requires knowing a possibly conservative bound on the Lipschitz constant of the system. We also provide a sample complexity result for the satisfaction of the specification on the system in closed loop with the designed controller for a given confidence. Our data-driven synthesised controller can provide guarantees on satisfaction of both finite and infinite-horizon specifications. We show that our data-driven approach can be readily used as a model-free abstraction refinement scheme by modifying the formulation of the system’s growth bounds and providing similar sample complexity results. The performance of our approach is shown on three case studies.

Sampled-data controller synthesis using dissipative linear periodic jump-flow systems with design applications

Sampled-Data Controller Synthesis using Dissipative Linear Periodic Jump-Flow Systems with Design Applications

Controller Design for Asynchronous Multirate Sampled-Data Systems

H∞ Controller Design for Asynchronous Hybrid Systems with Multiple Delays

Stability analysis and controller synthesis for discrete linear time-delay systems with state saturation nonlinearities

Sampled‐data MPC for LPV Systems with Input Saturation

Robust Extended Dissipative Control for Sampled-Data Markov Jump Systems.

Sampled-Data Control for Time-Delay Uncertain Linear Systems

On stability and H∞ control synthesis of sampled‐data systems: A multiple convex function approximation approach

Robust Sampled-Data Control for Uncertain Nonlinear Systems

Direct Data-Driven State-Feedback Control of Linear Parameter-Varying Systems

Mixed H∞ /passive Sampled-Data Synchronization Control of Complex Dynamical Networks with Distributed Coupling Delay

Model Predictive Control for Sampled-Data Systems

Periodic Sampled-Data-based Dynamic Model Control of Switched Linear Systems.

Sampled-Data Control for Asynchronously Switched Linear Systems Without MDT Constraints

Data-driven abstraction-based control synthesis

Data-driven control for discrete-time piecewise affine systems

Delay-dependent Dissipative Control for Linear Time-Delay Systems

A New Sampled-Data Output-Feedback Controller Design of Nonlinear Systems Via Fuzzy Affine Models

Sampled-Data Control for Linear Systems On Hybrid Time Domains

Controller Synthesis of Continuous-Time Piecewise Linear Systems Based on Piecewise Lyapunov Functions